Efficient and robust coupling methods for electro-mechanic models of the human heart
PIs: Prof. Christian Wieners, Prof. Axel Loewe
Aim:
This project aims to extend the limits of multi-physics simulations of the human heart by developing and evaluating robust coupling schemes for models of the cardiac electro-mechanical system with high biophysical accuracy across multiple scales and dimensions. By combining our profound interdisciplinary experience in modeling each of the cardiac physical systems separately with our preliminary work in coupling them, we are well positioned to construct and analyze an efficient and accurate parallel finiteelement realization of the fully coupled system.
As specific goals, we will develop and provide
(1) a formulation for a robustly coupled model for the whole heart comprising active electrophysiology, cardiac mechanics and circulatory function;
2) an efficient and scalable numerical implementation of this model and a comprehensive convergence analysis of the approximations in space and time;
(3) a user-friendly integration of the fully coupled model into the established openCARP simulation framework to foster community adoption of this research tool and pave the way for clinical translation.
Model equations ordered in ascending spatial dimension. At the micro level, ion concentrations and resulting contraction forces are calculated using cell models. The propagation of the electrical stimulus is modeled using the monodomain equation and the deformation of the heart using the impulse balance and a so-called „active stress“ approach. The blood circulation is described using a closed 0D circulation model.
Components of the fully coupled model. The ion and sarcomere models describe the processes at cell level and are coupled via calcium concentrations. The monodomain model describes the electrical propagation of the stimulus across the organ using the finite element method and is linked to the ion models via the transmembrane voltage and the resulting ion currents across the cell membrane. The deformation is calculated using the solid mechanics finite element model based on the contraction force of the sarcomere models. Pressures and volumes within the atria and ventricles are determined using a circulatory model. The deformation of the mechanics model affects the electrophysiology via several mechano-electric feedback (MEF) mechanisms.Both the spatial and temporal resolution differs for the different model components, which is why stable and robust coupling methods are of great importance.
Description:
Background:
Cardiac diseases are the number one cause of death in Germany. Computational modeling of the cardiovascular system can help to understand the relevant mechanisms and help to tailor treatments for heart diseases. Although cardiac computational modeling has significantly advanced in the last decades, this is often limited to a single function such as electrophysiology, biomechanics, blood flow in the heart or the circulatory system. These single-physics approaches are very valuable research tools, but the interaction of cardiac functions is required to fully leverage the potential of in silico approaches.
Methods:
Our multi-physics model consists of several model components that describe the electromechanical contraction of the heart at different spatial scales. Ion concentrations within the cell are modeled via ion channels of the cardiomyocyte cell membrane. Based on the resulting calcium concentration, a contraction force is obtained using a sarcomere model. The activation of the myocardium is based on the electrical propagation across the organ, which – like the deformation of the heart – is calculated using the finite element method. A circulatory model is coupled to the deformation of the heart via the volumes and pressures of the atria and ventricles.
Building on our current work, we will further develop our multi-physics model by extending it with stable and efficient implicit-explicit time integration and finite element formulations and making the fully coupled system scalable. Subsequently, we will use this advanced model to study the effects of local electrical dyssynchrony and the application of cardiac resynchronization therapy in collaboration with our clinical partners. In addition, we will identify application scenarios where the highly complex full model can be reduced and quantify the resulting uncertainty.
Involved Institutions:
Karlsruhe Institute of Technology
Institute of Biomedical Engineering
Computational Cardiac Modeling
Karlsruhe Institute of Technology
Institute for Applied and Numerical Mathematics
Scientific Computing
Applicants:
Publications
2024
In: The Journal of Physiology, 2024.
In: GAMM Mitteilungen, Bd. 46, Ausg. 2, 2024, ISSN: 1522-2608.